Optimizing the Treatment of Diabetes Patients. Medically complex patients present an important challenge for implementing quality measurement programs and for improving the quality of care in the U.S. healthcare system. Patients with type 2 diabetes are a growing population of complex patients with complicated treatment regimens and high number of comorbid conditions. Treatment prioritization in this population is increasingly important since patients with diabetes have, on average, 3.5 comorbid conditions with hypertension and hyperlipidemia being the most common. There has been an increasing emphasis on evaluating provider performance for evidence-based diabetes care. However, these measures assume one-size fits all, and do not account for changing needs and preferences of the patients with age, disease state, and comorbidities. Thus, it is often difficult for primary care clinicians to prioritize interventions for diabetes patients that would be sensitive to patient preferences and at the same time respond to external quality measurement requirements. Over the next two years, we will bridge this gap by advancing knowledge regarding the optimal treatment of diabetes as measured by the patient's quality adjusted lifespan, adherence to treatment, the costs of treatment, and cost of diabetes-related complications to the health system. Specifically, we will focus on developing mathematical models to determine the optimal time for initiating and intensifying treatment for the management of diabetes, hypertension, and hyperlipidemia. Using longitudinal datasets, we aim to better understand the medication taking behavior in diabetes patients, most importantly the aspects related to adherence. The results of these analyses will then be incorporated into our mathematical models. Thus, our models will provide optimal timing of treatment assuming perfect adherence, but also at different levels of non-adherence. Incorporating non- adherence will provide us more realistic scenarios related to treatment and greater opportunity to influence the patient's length and quality of life. The findings from this study will then be translated to decision aids. These decision aids can be tested in practical randomized trials to evaluate the extent to which they can realize improved value of healthcare by enhancing outcomes and quality of care at an efficient cost. [unreadable] [unreadable] [unreadable] [unreadable]